5 Things to Get Right Before You Spend a Dollar on Ads

Ads are the easiest part of marketing to turn on, and the easiest part to waste money on.

That isn't because ads don't work. It's because an ad only does one thing: it buys a moment of someone's attention and sends them somewhere. Everything that happens after the click was decided before you spent a dollar. The page they land on. The offer they see. How fast someone answers them. Whether you can even tell which ad brought them in.

Ads don't fix a business. They multiply it. If what's behind the ad works, you grow faster. If it doesn't, you find out faster, and you pay for the lesson.

I've built lead generation for restaurants, luxury retail brands and medical businesses, and I've sent more than 1,000,000 cold emails through outreach systems I built. Different customers, different channels, same rule: volume never fixes a weak foundation. It only exposes it sooner.

So before you spend on ads, get these five things right. They're in order, because each one depends on the one before it.

1. A position a customer could repeat back to you

Every ad borrows from your positioning. The headline, the image, the audience you target, the page you send people to. If the position is vague, the ad is vague, and a vague ad competes on budget alone. Someone will always have a bigger budget than you.

Positioning answers three questions:

• Who exactly is this for? Not "small businesses" or "homeowners." The person most likely to buy, most likely to be happy, and most likely to send you someone else.
• What do you do for them that they care about? The result, in their words, not yours.
• Why you instead of the alternatives? Including the one most businesses forget: doing nothing.

If your answer leans on "quality," "great service" or "years of experience," you don't have a position yet. Every competitor says those things, so they do no work.

Here's a quick test. Cover your logo on your ad and on the page it sends people to. Could a competitor run the exact same thing tomorrow? If so, you're paying to advertise your category, not your business.

The biggest lesson from a million cold emails applies directly here. The difference between outreach that works and outreach that doesn't is almost never the clever line. It's who gets the message, and whether the message is plainly about them. Ads behave the same way. A narrow audience and a specific message beat a broad audience and a perfect headline.

Before you write a single ad, write one sentence:

For (who), who (situation), we (result), unlike (the alternative), because (proof).

If you can't fill in the proof, that's the first thing to fix. No ad budget will fix it for you.

2. An offer that gives a stranger a reason to act

"Contact us" isn't an offer. It's a request, and it asks a stranger to do work with nothing in it for them.

An offer answers the question every visitor is silently asking: what do I get, and what does it cost me? Cost isn't only money. It's time, effort, and the risk of being sold to.

The best first offers are useful on their own, even if the person never buys. Three I've built:

• A free assessment. Dr. Bertrand Desilva is a board-certified sleep physician, and I built his website, CRM and social media. The site gives people two ways in: book a telehealth visit, or take the free "Am I at Risk?" sleep assessment first. Not everyone worried about their sleep is ready to see a doctor today. The assessment gives them a step they are ready for.
• An instant answer. Buy Sell Gold San Diego, a family-run precious-metals and estate buyer in the Gaslamp Quarter, has an instant value calculator for sellers. The first question anyone selling gold has is "what is it worth?" The calculator answers it before they have to call anyone.
• Something specific to buy. Buon Cibo Italian Specialties, a deli and catering business in Scripps Ranch that I built through Clear the Plate, my restaurant consulting company, takes prepaid preorders for dinners, weekly specials and heat-and-eat meals, plus tickets for wine nights and events. So every email and every post has something concrete to point to. The website brings in about $20,000 a month in prepaid orders, separate from the counter.

Match the offer to how warm the visitor is. Someone who searched for exactly what you sell is warm, and may be ready to book. Someone who saw your ad while scrolling is cold; five seconds ago they weren't thinking about you at all. Cold traffic needs a smaller first step: an assessment, a calculator, a guide, a first order. Asking a cold visitor to book a sales call is asking too much, too soon.

Run your offer through these questions:

• Can someone understand what they get in one sentence?
• Is it useful even if they never buy from you?
• Is the first step small enough for someone who has never heard of you?
• Is there a reason to do it now instead of "someday"?
• Do you know exactly what happens after they say yes?

3. A page built for the click, not your homepage

Most businesses send ad traffic to their homepage. It's almost always the wrong place.

Your homepage has to serve everyone: the referral checking you out, the existing customer looking up your hours, the person who searched your name. The ad visitor arrived with one specific promise in mind. Send them to a page built for that promise and nothing else.

What that page needs:

• Message match. The headline continues the ad. If the ad offered a free assessment, the headline is about the free assessment, not "Welcome to" your company. The visitor should know in a second that they landed in the right place.
• One audience, one action. One primary button, repeated down the page. Remove the main navigation and any link that leads somewhere else. Every exit is a click you paid for walking out the door.
• Proof near the top. Specific beats impressive. A real number, a real credential, a real review with a real name.
• Answers to the objections. What it costs or how pricing works, how long it takes, what happens next, who they'll talk to.
• A form that asks only for what you need. Every extra field is one more reason to stop.
• A phone-first build. Open the page on your phone, on cellular data, with one thumb. That's how a lot of the people you're paying for will see it.

In high-trust categories the page has to look as serious as the purchase. I've built more than 70 websites for luxury watch dealers, every one of them by referral, and many of those dealers sell five- and six-figure pieces without a storefront. Online, they have to look as established as their inventory, because in that category the first impression is most of the sale. If your ad sends someone who's about to spend real money to a page that looks thrown together, the ad did its job and the page undid it.

If you're not sure your site is ready for paid traffic, I wrote about how to tell if your website is costing you leads. Building pages that turn traffic into customers is what Websites & Funnels is for.

4. Tracking that follows the lead all the way to the sale

If you can't see which ad produced which customer, you're not running a campaign. You're guessing with a budget.

Start by deciding what counts. For each campaign, pick the one action that matters, and the money event behind it:

• A restaurant: a preorder, a reservation, or an event ticket.
• A luxury retailer: an inquiry about a specific piece, or a booked appointment.
• A medical practice: a completed assessment, then a booked consultation.

Then put the minimum in place:

• Analytics on every page, with that action set up as a conversion.
• The ad platform's own conversion tracking, installed and tested, whether that's the Google Ads conversion tag or the Meta pixel.
• UTM tags on every ad link, so each visit carries its source with it.
• Call tracking if your leads phone you. A call your analytics can't see is a lead with no source.
• A source field on every lead in your CRM, filled in automatically, so the sale can be traced back to the ad.

This matters for more than reporting. Ad platforms optimize toward whatever you tell them success is. Tell them a page view is success, and they'll find you people who view pages. Tell them a booked appointment is success, and they'll go looking for people who book.

Before launch, test it end to end. Submit your own form. Make a test call. Place a test order. Confirm each one shows up in your analytics, in the ad platform and in your CRM, with the right source attached. It's tedious, and it's the only way to know the numbers you'll be making decisions on are real.

Then measure past the click: cost per lead, booking rate, show rate, close rate, cost per sale, and revenue by campaign. A cheap lead that never buys is the most expensive lead you have.

5. Follow-up that answers before the lead goes cold

This is where most of the money leaks, and it's the leak nobody sees.

Someone fills out your form at nine on a Saturday night. What happens next?

For most businesses the honest answer is that someone gets to it eventually. Maybe in an hour. Maybe Monday. Maybe never. Meanwhile that person is still on their phone, and your competitor's ad is the next thing they see.

You don't need a statistic to know speed matters. Think about the last time you reached out to three businesses at once. The one that answered first usually got the conversation.

Answer these before you launch:

• Who gets notified when a lead comes in, and how?
• How fast does the first response go out, including nights and weekends?
• Does that first response move them to a next step, or just say "we got your message"?
• How many times, over how many days, do you follow up with someone who doesn't reply?
• Where do people go who aren't ready yet? Onto an email list, or nowhere?
• Who books the appointment, and who handles reschedules and no-shows?

You can do all of this by hand if you truly commit to it. But manual follow-up is the first thing that slips when the business gets busy, and the business gets busy exactly when your ads start working. That's why I build follow-up as a system.

At Solve Sleep Apnea, a nationwide sleep apnea network I co-founded, AI agents, including voice agents, handle patient calls, texts and email. They qualify and route patients 24/7, and they set and reschedule appointments. The team runs the consultations and routes each patient to the right partner provider. The system handles the rest. Someone who reaches out at midnight gets answered at midnight.

Every lead you pay for and don't answer quickly is a lead you bought for your competitor.

I wrote more about what good automated follow-up looks like, and what bad follow-up sounds like. It's also the core of what I build in Business Automations.

The order matters more than the budget

Position, offer, page, tracking, follow-up. Then ads. Each piece depends on the one before it:

• Without a position, the offer is generic.
• Without an offer, the page has nothing to ask for.
• Without a page built for the click, the traffic has nowhere useful to go.
• Without tracking, you can't tell what's working, and neither can the ad platform.
• Without follow-up, the leads you paid for go cold.

When you do turn ads on, treat the first budget as the cost of learning, not the engine of growth.

Pick one channel to start. Choose it by where your customer is. Google Ads reach people who are searching for what you sell right now. Meta ads reach people who fit your customer but aren't searching yet, which takes stronger creative and a longer follow-up.

Set a test budget you're comfortable learning from. Not the budget you hope to scale to. The budget that buys you an honest answer.

Decide your rules before you launch. Write down what result would make you stop, what would make you change something, and what would make you spend more. It's much easier to make that call before the money is spent than after.

Build an audience you own alongside the one you rent. Buon Cibo opened with more than 1,500 email subscribers and more than 3,000 social followers, because the work started about a year before the doors opened. A list you own is the cheapest place to test an offer: you can try it on people who already know you before you pay to show it to strangers.

The pre-launch checklist

Before the first dollar goes out, you should be able to say yes to every one of these:

1. I can describe who this is for in one sentence, and a real customer would agree with it.
2. A competitor couldn't run my ad unchanged.
3. My offer is useful even if the person never buys.
4. The first step is small enough for someone who has never heard of me.
5. Ad traffic goes to a page built for that ad, not to my homepage.
6. That page's headline matches the ad's promise.
7. The page works on a phone, with one thumb, on cellular data.
8. There's one primary action, and nothing on the page pulls people away from it.
9. I've tested a form, a call and, if it applies, a purchase, and each one shows up with its source.
10. Every lead gets a fast first response, including nights and weekends.
11. People who don't reply get a planned series of follow-ups, not a single try.
12. I've written down what result means stop, change or scale.

If you can't say yes to all twelve, you know what to fix first. Fixing it costs less than running ads on top of it.

Where to start

If you're about to spend on ads, or you're already spending and can't tell what's working, start underneath the ads.

That's why every engagement with me starts with a Strategic Blueprint, before anything goes live. For a business about to advertise, it's where we find the leaks between the click and the sale, and decide where the budget should go first. It usually takes about 10 business days, anywhere from 2 to 14 depending on the business. I wrote more about why the blueprint comes before the website, and the same logic applies to ads.

When the foundation is ready, Marketing & Advertising is where I build and run the campaigns, measured against revenue, not clicks. If you'd like a second set of eyes before you spend, get in touch. Tell me where your customers come from now and what you've tried, and I'll tell you where I'd look first.